[GH-ISSUE #3981] Question regarding KoboldCPP integration #2532

Closed
opened 2026-02-22 18:30:06 -05:00 by yindo · 1 comment
Owner

Originally created by @CrazyAce25 on GitHub (Jun 11, 2025).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/3981

Originally assigned to: @shatfield4 on GitHub.

I am attempting to use KoboldCPP as the back end for AnythingLLM and while I am able to get it to connect and utilize basic chat functionality, I am unable to use the vision or stable diffusion capabilities provided by KoboldCPP. LM Studio also provides vision capabilities, but Ive likewise been unable to utilize this within AnythingLLM. Is there something more I need to do to enable this functionality or is this something that doesn't currently function?

Originally created by @CrazyAce25 on GitHub (Jun 11, 2025). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/3981 Originally assigned to: @shatfield4 on GitHub. I am attempting to use KoboldCPP as the back end for AnythingLLM and while I am able to get it to connect and utilize basic chat functionality, I am unable to use the vision or stable diffusion capabilities provided by KoboldCPP. LM Studio also provides vision capabilities, but Ive likewise been unable to utilize this within AnythingLLM. Is there something more I need to do to enable this functionality or is this something that doesn't currently function?
yindo closed this issue 2026-02-22 18:30:06 -05:00
Author
Owner

@shatfield4 commented on GitHub (Jun 11, 2025):

After testing this with a vision model, I can confirm that it is working properly in AnythingLLM. I used the vision model shown in this video to test it out and was able to get a response back in AnythingLLM describing images I uploaded to the chat. Once you have your vision model loaded and have KoboldCPP hooked up to AnythingLLM, you can upload images into the AnythingLLM chat window and ask it questions about the image.

In regards to the stable diffusion capabilities, at the current moment, we do not support image generation inside AnythingLLM. This will be something that we may explore in the future though.

@shatfield4 commented on GitHub (Jun 11, 2025): After testing this with a vision model, I can confirm that it is working properly in AnythingLLM. I used the vision model shown in [this video](https://www.youtube.com/watch?v=U4ekEOBc2LU&ab_channel=FahdMirza) to test it out and was able to get a response back in AnythingLLM describing images I uploaded to the chat. Once you have your vision model loaded and have KoboldCPP hooked up to AnythingLLM, you can upload images into the AnythingLLM chat window and ask it questions about the image. In regards to the stable diffusion capabilities, at the current moment, we do not support image generation inside AnythingLLM. This will be something that we may explore in the future though.
yindo changed title from Question regarding KoboldCPP integration to [GH-ISSUE #3981] Question regarding KoboldCPP integration 2026-06-05 14:47:05 -04:00
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Mintplex-Labs/anything-llm#2532